Noise, Vibration and Harshness (NVH) is a comprehensive issue of manufacturing quality, it gives the purchaser the most direct and superficial feeling. With economic development, the improvement of people's living standards, as well as the needs of the export market, it is urgent to develop and produce some better NVH performance motorcycles.

The vibration and noise of the motorcycle have a great impact on people and are the core issues of NVH. Some countries have advanced motorcycle design and manufacturing technologies, and developed advanced software and hardware for comfort analysis and the research on the sound quality of motorcycles, which established general guidelines for human comfort. At present, it is a common solution, basically starting from the transmission path of the vibration and noise, to solve the vibration and noise of motorcycles.^1-5

Transmission error (T.E.) is the most important factor of gear noise and vibration because transmission errors affect the changes of the force and the stiffness of gears.^6,7 In this paper, the transmission of the scooter gearbox is modeled and analyzed by RomaxDesigner, the 1st helical gear pair of the gearbox has been investigated through static analysis, and through gear micro-modification, the gear pair and overall noise is reduced by minimizing gear PPTE. At last, simulation results of gear micro-modification are confirmed by noise and vibration measurement on bench testing. It was determined that the optimum gear micro-modification could be applied to minimize the transmission error and reduce the overall noise of the scooter gearbox.

Scooter transmission system mainly consists of four components: drive shaft, intermediate gear, intermediate shaft and final gear assembly, and the schematic of the gearbox is shown in Fig. 1.

Fig. 1 
The schematic of the scooter gearbox

The speed of the scooter engine is transmitted through the belt between the driving pulley of CVT (Continuously Variable Transmission) on the crankshaft and the driven pulley on the drive shaft, different gear ratios are obtained by the changing diameters of the driving pulley and the driven pulley.

Engine power is driven by the belt to the drive shaft, and through the drive shaft meshing with the intermediate gear, and the intermediate gear and the intermediate shaft are fitted by involute spline, he intermediate shaft meshes with the final gear assembly (The final gear and the output shaft are assembled by interference fit.) to transmit power to the rear wheel on the output shaft to drive the scooter.

As the speed of scooter’s rear wheel is nearly up to thousands of revolutions, so these gears are high-speed rotating gears. Besides the CVT is mainly used for the engine in the scooter, due to large changes in transmission ratio and speed, the gear meshing between the driving gear of the input shaft and the intermediate gear (ESP. 1st gear pair) is the main part of the noise in this transmission system.

2.1 Scooter Noise

Scooter noise, according to the nature of its generation, is roughly divided into three categories: mechanical noise, aerodynamic noise and electromagnetic noise.

All of these are resulting from the interaction of the relevant components. The relationship of their composition is shown in Fig. 2.

Fig. 2 
The relationship of the scooter noise

The basic method to reduce the noise of the scooter is to control the main source of noise. Therefore, in order to control scooter noise, the noise source is firstly needed to be located to find the main sound source.

In order to inspect the effect of gear profile micro-modification and reduce gearbox noise, a scooter gearbox without case housing (The impact of gearbox housing stiffness is not considered.¹³) is modeled in RomaxDesigner software to investigate the T.E. under the design torques, which G is the gravity of CVT driven pulley and F is the tension force of CVT belt. Fig. 3 is the modeling of a scooter gearbox without housing, Fig. 4 it the 1st helical gear pair, and Table 1 is the summary of the specification of the 1st gear pair of the gearbox.

Fig. 3 
Modeling of a scooter gearbox without housing

Fig. 4 
The 1st helical gear pair

As we known, micro-modification parameters cannot be varied individually to evaluate the sensitivity of the durability and NVH performance, and even ideal designs can rarely be manufactured for mass production. But in order to quickly evaluate the results of the nominal modification, esp. profile tip relief, only nominal values of gear micro-modification are used for this gear pair without the consideration of manufacturing tolerance. And the nominal modification parameters are given in Table 2. The typical input loaded torques are 10.79 Nm, 7.46 Nm and 4.79 Nm which are named as Load Case 1 (LC1), Load Case 2 (LC2) and Load Case 3 (LC3) respectively. By simulation, PPTE before and after gear micro-modification is obtained as shown in Fig. 5. Transmission error is reduced at all load cases. And contact patter of one load case before and after modification is shown in Fig. 6.

Fig. 5 
PPTE before and after micro-modification

Fig. 6 
Contact pattern before and after micro-modification

Along the tooth width direction, left tooth edge is loaded with more loads compared to right edge, which is optimized so that both edge receives balanced load. The contact load distribution shows that the optimized has avoided edge contact, and the load is distributed over a wide area of the tooth, maximum contact load is close to the center of gear width.

A design with very good performance may give inferior performance in experimental conditions. It is important to correlate the simulation results with actual experimental results of a gear pair to confirm accuracy of transmission model and analysis condition during CAE. In order to acquire gear noise and vibration data, the gearbox is tested in a semi-anechoic room as shown in Fig. 7. Noise and Vibration data was acquired by B&K testing software in run-up and run-down condition without specific torque applied at the input shaft of the gearbox (Scooter purchasers always test in situ which means the load of the drive shaft is small.). Example comparison for 1st gear pair and overall in run-up noise is illustrated in Fig. 8. It shows that the noise level is improved in dB for 1st gear pair and overall after gear micro-modification. It is possible to conclude that simulation results of gear micro-modification are confirmed by noise and vibration measurement on bench testing.

Fig. 7 
The tested gearbox in a semi-anechoic room

Fig. 8 
Comparison of gear noise in 1st gear pair and overall before and after gear micro-modification

Time and cost are two critical parameters for the transmission development, which means durability, efficiency and NVH performance must be confirmed as early as possible during production development. In this paper, the transmission of the scooter gearbox is modeled and analyzed by RomaxDesigner, the 1st helical gear pair of the gearbox has been investigated through static analysis, and through gear micro-modification, the gear pair and overall noise is reduced by minimizing gear PPTE. The comparison of the simulation to prototype testing shows good correlation. Traditional mode of experimentation with repeated prototyping can be avoided, which means simulation optimization can greatly improve the research and development efficiency of the transmission system.